Table 5. Cox proportional hazard modelsa for the risk of dying associated with the use of current hormone treatment (versus non-current use) at baseline, stratified by estrogen receptor genotype.
| Multivariate-adjustedb risk of mortality associated with current HT | ||||||
| Genotype | All-cause | Cancer-related | ||||
| Deaths | HR (95% CI) | p | Deaths | HR (95% CI) | p | |
| ESR1 rs2234693 | ||||||
| Interaction term | 0.006 | 0.004 | ||||
| TT (n = 1351) | 85 | 1.65 (0.77–3.55) | 0.20 | 32 | 3.18 (1.23–8.20) | 0.017 |
| CC/CT (n = 3112) | 196 | 0.42 (0.18–0.97) | 0.042 | 70 | 0.38 (0.12–1.27) | 0.12 |
| ESR1 rs9340799 | ||||||
| Interaction term | 0.030 | 0.006 | ||||
| AA (n = 1870) | 115 | 1.39 (0.68–2.82) | 0.37 | 41 | 2.43 (1.04–5.70) | 0.040 |
| GG/GA (n = 2593) | 166 | 0.42 (0.17–1.04) | 0.059 | 61 | 0.31 (0.07–1.32) | 0.31 |
| ESR2 rs1271572 | ||||||
| Interaction term | 0.060 | 0.019 | ||||
| GG (n = 1434) | 93 | 1.59 (0.76–3.35) | 0.22 | 37 | 2.90 (1.20–7.03) | 0.018 |
| TT/TG (n = 3029) | 188 | 0.43 (0.19–0.99) | 0.049 | 65 | 0.40 (0.12–1.32) | 0.13 |
a
This analysis was carried out on a sub-population of 4463 women for whom genotyping data was available.
b
Adjusted for age, education, living status, recruitment centre, incapacities, comorbidity, depressive symptoms and cognitive impairment.